This proposal is in response to the NIH call for Exploratory Collaborations with National Centers for Biomedical Computing, PAR-06-223, and it will involve a collaboration between Columbia University's MAGNet NCBC and a team at Los Alamos National Laboratory. The aim of the proposal is a system-wide tudy of integrated transcriptional and metabolic networks in Eschericia coli K-12 strain, aiming at a similar analysis of a pathogen, Bacillus anthracis, at a later date. LANL hosts an experimental research program on bacterial metabolomics. Metabolites serve several functions. The most common one is being the precursors to various cellular components. They are also regulators of cellular functions by means of modulating metabolic reactions or binding to transcription factors and subsequently regulating gene expression. Conversely, the genes regulated by a transcription factor often encode enzymes, modulating the speed of metabolic reactions. Thus, to understand and ultimately predict the cellular response to an environmental change of interest (e.g., pathogen entry into its host environment), we must integrate the analysis of the transcriptome and metabolome. To address this need, we will work with the laboratories of Pat Unkefer and John Dunbar, which will produce data sets of about 300 joint metabolic/transcriptional profiles of E.coli under different steady-state growth conditions. The resources of MAGnet NCBC, specifically the algorithms within the geWorkbench bioinformatics platform produced by the center, will be leveraged to reconstruct cellular networks. Specifically, we expect that ARACNE, an algorithm originally developed by MAGNet for high-fidelity analysis of transcriptional networks in mammalian cells, is well positioned for reconstruction of metabolic networks from high throughput system-wide metabolic activity data, provide that appropriate modifications to deal with the specifics of the metabolic data are made. We will also adapt the algorithm to discover modulated interactions, that is, metabolic interactions that are conditional on the activity of a modulator gene (enzyme), or transcriptional interactions that require the presence of a metabolite to proceed. Such integrated genome/metabolome analysis has not been attempted yet. It will be a giant leap towards a complete understanding of cellular processes in an important organism. Because of the comparatively small size of bacterial genomes and metabolomes, it will be possible to perform system-wide analyses of interactions for the entire integrated genome and metabolome. While important in its own right, especially in view of the pathogenic nature of B. anthracis, this research would also represent an important test bed for a subsequent study of metabolic diseases in higher animals, including humans. [unreadable] [unreadable] [unreadable]